Inspiration
We’ve all opened a dataset and thought, “What the hell am I looking at?” so i built something that explains it for you- fast.
What it does
You upload a CSV, it gives you an AI-generated summary, and you can search those insights later using natural language. It tells you the type of data, highlights quality issues, and suggests cleaning steps and visualizations including which charts or plots would be most useful for different columns
How we built it
React frontend, FastAPI backend, used Gemini for analysis(no gcp credits unfortunately🥲), MongoDB Atlas for storing vectors, and a clean vector search setup.
Challenges we ran into
Getting vector search to play nice with MongoDB was trickier than expected but overall was fun building it.
Accomplishments that we're proud of
The AI summaries are actually useful, the search works like magic and it’s all stitched together in a simple UI
What we learned
Fast iteration with AI and databases takes finesse- but when it clicks, it really clicks
What's next for NarratAIve Data Analyzer
Multi-file uploads, visualizations for each dataset, smarter trend detection, and turning it into a full-blown AI-powered data assistant not just a smart notetaker
Log in or sign up for Devpost to join the conversation.